Sungwon Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sungwon Lee has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Dongkyun Kim
School of Computer Science and Engineering, Kyungpook National University, Daegu, Republic of Korea
3 shared publications - Jung Min Cho
Functional Food Research Institute, Industry-University Cooperation Foundation, Daegu Haany University, Gyeongsan-si 38610, Republic of Korea
2 shared publications - Junho Seo
Kyungpook National University, School of Computer Science and Engineering, South Korea
2 shared publications - Soo Yeon Lee
Department of Cosmetics, Daegu Haany University, Gyeongsan, Korea
2 shared publications - Ayesha Siddiqa
Wireless and Mobile Internet Laboratory, School of Computer Science and Engineering, Kyungpook National University, Daegu, South Korea
1 shared publication - Bomi Jeong
Kyungpook National University, School of Computer Science and Engineering, South Korea
1 shared publication
Latest Publications
- Open Access
- 31 October 2026
- Published by Korean Society of Clinical Nutrition in Clinical Nutrition Research
- Vol. 15 (4) , 269-276
Development of an institutional framework based on locally deployed large language models for clinical nutrition research: from fragmented documents to reusable knowledge
Clinical nutrition studies generate diverse research materials, including study protocols, case report forms, dietary records, laboratory measurements, body composition data, and final reports. However, these resources are often fragmented across storage systems and remain underutilized after study completion. Large language models (LLMs) offer opportunities for knowledge retrieval and documentation, but cloud-based services raise concerns regarding data privacy, institutional governance, and regulatory compliance. This Research Note proposes an institutional local LLM-based framework for secure knowledge retrieval, research documentation, and knowledge reuse in clinical nutrition research. The framework integrates multidisciplinary perspectives from clinical nutrition, human intervention trials, and artificial intelligence. Its architecture comprises five interconnected layers: a clinical nutrition data layer encompassing research documents and structured datasets; a document processing layer for text extraction, metadata generation, and standardization; a knowledge repository layer combining structured and vector databases for metadata management and semantic retrieval; a local LLM layer integrating retrieval-augmented generation with a separate numerical analysis pipeline; and a researcher interface supporting natural language queries and evidence-linked documentation. The proposed framework provides a conceptual foundation for secure, reusable institutional research knowledge management.
- 31 October 2026
- Published by Korean Society of Clinical Nutrition in Clinical Nutrition Research
- Vol. 15 (4) , 269-276
Open AccessDevelopment of an institutional framework based on locally deployed large language models for clinical nutrition research: from fragmented documents to reusable knowledge
Clinical nutrition studies generate diverse research materials, including study protocols, case report forms, dietary records, laboratory measurements, body composition data, and final reports. However, these resources are often fragmented across storage systems and remain underutilized after study completion. Large language models (LLMs) offer opportunities for knowledge retrieval and documentation, but cloud-based services raise concerns regarding data privacy, institutional governance, and regulatory compliance. This Research Note proposes an institutional local LLM-based framework for secure knowledge retrieval, research documentation, and knowledge reuse in clinical nutrition research. The framework integrates multidisciplinary perspectives from clinical nutrition, human intervention trials, and artificial intelligence. Its architecture comprises five interconnected layers: a clinical nutrition data layer encompassing research documents and structured datasets; a document processing layer for text extraction, metadata generation, and standardization; a knowledge repository layer combining structured and vector databases for metadata management and semantic retrieval; a local LLM layer integrating retrieval-augmented generation with a separate numerical analysis pipeline; and a researcher interface supporting natural language queries and evidence-linked documentation. The proposed framework provides a conceptual foundation for secure, reusable institutional research knowledge management.
- Open Access
- 23 October 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e103706
Data-Rich But Integration-Poor Human Trials: Conceptual Framework for Closing the AI Transformation Gap in Dietary Intervention and Cosmetic Human Trials
Human trials in dietary and cosmetic domains generate large volumes of multimodal data but remain limited in their ability to produce integrated and actionable insights. Data such as intake records, physiological biomarkers, imaging outputs, and user-reported outcomes are typically collected through disconnected systems and analyzed independently, restricting the ability to establish time-resolved relationships between exposure and response. In this viewpoint, we define this limitation as the AI transformation (AX) gap, representing a structural disconnect between data generation and actionable knowledge in human trials. Drawing on evidence from real-world industry demand, we highlight how this gap is consistently observed across dietary intervention trials, cosmetic human trials, and underlying data systems. We propose a conceptual AX-driven framework that integrates multimodal data acquisition, continuously aligns intervention exposure with corresponding physiological, phenotypic, and participant-reported responses across time, and supports centralized analytical processing. The framework is intended to provide a unified perspective for transforming fragmented human-trial data into integrated, interpretable, and actionable knowledge. Although the proposed framework has not yet been empirically validated, it offers a conceptual foundation for future development and evaluation of integrated data platforms in dietary intervention and cosmetic human trials. Closing the AX gap may facilitate the transition from fragmented data collection toward human trials in which multimodal exposure and response data are continuously linked across time to support more integrated and individualized interpretation in research and product development.
- 23 October 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e103706
Open AccessData-Rich But Integration-Poor Human Trials: Conceptual Framework for Closing the AI Transformation Gap in Dietary Intervention and Cosmetic Human Trials
Human trials in dietary and cosmetic domains generate large volumes of multimodal data but remain limited in their ability to produce integrated and actionable insights. Data such as intake records, physiological biomarkers, imaging outputs, and user-reported outcomes are typically collected through disconnected systems and analyzed independently, restricting the ability to establish time-resolved relationships between exposure and response. In this viewpoint, we define this limitation as the AI transformation (AX) gap, representing a structural disconnect between data generation and actionable knowledge in human trials. Drawing on evidence from real-world industry demand, we highlight how this gap is consistently observed across dietary intervention trials, cosmetic human trials, and underlying data systems. We propose a conceptual AX-driven framework that integrates multimodal data acquisition, continuously aligns intervention exposure with corresponding physiological, phenotypic, and participant-reported responses across time, and supports centralized analytical processing. The framework is intended to provide a unified perspective for transforming fragmented human-trial data into integrated, interpretable, and actionable knowledge. Although the proposed framework has not yet been empirically validated, it offers a conceptual foundation for future development and evaluation of integrated data platforms in dietary intervention and cosmetic human trials. Closing the AX gap may facilitate the transition from fragmented data collection toward human trials in which multimodal exposure and response data are continuously linked across time to support more integrated and individualized interpretation in research and product development.
- Open Access
- 6 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Experimental Analysis of Handover Process in Cell-Free Networks
With the emergence of services that require a high level of data usage, wireless networks are required to support higher data rates and greater capacity. To satisfy these requests, 5G architecture supports a high-frequency band. However, a high frequency occurs small range resulting in a need for numerous base stations (BSs). Deploying numerous BSs can be considered practically impossible due to high costs. To address the cost issue, instead of BSs, access points (APs) are used in Cell-free networks. That is, Cell-free networks that offer promising coverage gain and enhanced data rates consist of multiple APs and a single central process unit (CPU). This coverage of multiple APs' is much smaller than BS's. As a result, this small coverage can lead to frequent handovers. To overcome this frequent handover, an optimized handover scheme for the Cell-free network is required. Despite this optimization requirement, many studies in the Cell-free network field remain at the physical layer stage. No basic handover process is optimized for Cell-free networks, and no protocol for upper layers is defined. We consider that existing technologies will continue to be used in Cell-free networks, even if the protocols optimized for Cellfree networks are developed and used. For these reasons, we construct a Cell-free network architecture similar to the 5G structure for the handover process. In this paper, we investigate the handover process in 5G architecture and extend it to a Cellfree network environment. In our simulations, APs initiate and perform the handover process based on measurement reports from users. We analyzed and discussed the performance matrices of the handover in a Cell-free network environment, such as handover delay and throughput.
- 6 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessExperimental Analysis of Handover Process in Cell-Free Networks
With the emergence of services that require a high level of data usage, wireless networks are required to support higher data rates and greater capacity. To satisfy these requests, 5G architecture supports a high-frequency band. However, a high frequency occurs small range resulting in a need for numerous base stations (BSs). Deploying numerous BSs can be considered practically impossible due to high costs. To address the cost issue, instead of BSs, access points (APs) are used in Cell-free networks. That is, Cell-free networks that offer promising coverage gain and enhanced data rates consist of multiple APs and a single central process unit (CPU). This coverage of multiple APs' is much smaller than BS's. As a result, this small coverage can lead to frequent handovers. To overcome this frequent handover, an optimized handover scheme for the Cell-free network is required. Despite this optimization requirement, many studies in the Cell-free network field remain at the physical layer stage. No basic handover process is optimized for Cell-free networks, and no protocol for upper layers is defined. We consider that existing technologies will continue to be used in Cell-free networks, even if the protocols optimized for Cellfree networks are developed and used. For these reasons, we construct a Cell-free network architecture similar to the 5G structure for the handover process. In this paper, we investigate the handover process in 5G architecture and extend it to a Cellfree network environment. In our simulations, APs initiate and perform the handover process based on measurement reports from users. We analyzed and discussed the performance matrices of the handover in a Cell-free network environment, such as handover delay and throughput.
- Open Access
- 25 August 2020
- Published by Wiley in Wireless Communications and Mobile Computing
- Vol. 2020, 1-9
Avoiding Spurious Retransmission over Flooding-Based Routing Protocol for Underwater Sensor Networks
In underwater wireless sensor networks (UWSN), acoustic communication naturally introduces challenges such as long propagation delay and high packet loss. The flooding-based routing protocol can address these challenges with its multipath characteristics. As in flooding-based routing, due to multipath propagation mechanism, not only DATA but also ACK messages are transmitted through multiple routes however still some packet loss will degrade the performance. So, to provide high reliability of message delivery, an efficient retransmission mechanism is inevitable. Though, if the network uses conventional transport layer protocol such as TCP, it will suffer a spurious retransmission problem as TCP was originally not designed for the multipath environment. In this paper, we propose route discrimination for flooding-based routing to reduce spurious retransmission in UWSN to solve the limitation. The notion of ACK copies waiting time (ACWT) is utilized which is selectively updated based on the similarity of paths of transmission of ACK message copies. We also improved our previous solution that lacks flexibility to cope with dynamic link error characteristics. Through evaluation, we verified that our new scheme achieves the performance improvements of 14%~84% in terms of retransmission ratio compared to the previous research.
- 25 August 2020
- Published by Wiley in Wireless Communications and Mobile Computing
- Vol. 2020, 1-9
Open AccessAvoiding Spurious Retransmission over Flooding-Based Routing Protocol for Underwater Sensor Networks
In underwater wireless sensor networks (UWSN), acoustic communication naturally introduces challenges such as long propagation delay and high packet loss. The flooding-based routing protocol can address these challenges with its multipath characteristics. As in flooding-based routing, due to multipath propagation mechanism, not only DATA but also ACK messages are transmitted through multiple routes however still some packet loss will degrade the performance. So, to provide high reliability of message delivery, an efficient retransmission mechanism is inevitable. Though, if the network uses conventional transport layer protocol such as TCP, it will suffer a spurious retransmission problem as TCP was originally not designed for the multipath environment. In this paper, we propose route discrimination for flooding-based routing to reduce spurious retransmission in UWSN to solve the limitation. The notion of ACK copies waiting time (ACWT) is utilized which is selectively updated based on the similarity of paths of transmission of ACK message copies. We also improved our previous solution that lacks flexibility to cope with dynamic link error characteristics. Through evaluation, we verified that our new scheme achieves the performance improvements of 14%~84% in terms of retransmission ratio compared to the previous research.
- Open Access
- 28 May 2020
- Published by MDPI AG in Electronics
- Vol. 9 (6) , 902
Neighbor Aware Protocols for IoT Devices in Smart Cities—Overview, Challenges and Solutions
The principle of Smart Cities is the interconnection of services, based on a network of Internet of Things (IoT) devices. As the number of IoT devices continue to grow, the demand to organize and maintain the IoT applications is increased. Therefore, the solutions for smart city should have the ability to efficiently utilize the resources and their associated challenges. Neighbor aware solutions can enhance the capabilities of the smart city. In this article, we briefly overview the neighbor aware solutions and challenges in smart cities. We then categorize the neighbor aware solutions and discuss the possibilities using the collaboration among neighbors to extend the lifetime of IoT devices. We also propose a new duty cycle MAC protocol with assistance from the neighbors to extend the lifetime of the nodes. Simulation results further coagulate the impact of neighbor assistance on the performance of IoT devices in smart cities.
- 28 May 2020
- Published by MDPI AG in Electronics
- Vol. 9 (6) , 902
Open AccessNeighbor Aware Protocols for IoT Devices in Smart Cities—Overview, Challenges and Solutions
The principle of Smart Cities is the interconnection of services, based on a network of Internet of Things (IoT) devices. As the number of IoT devices continue to grow, the demand to organize and maintain the IoT applications is increased. Therefore, the solutions for smart city should have the ability to efficiently utilize the resources and their associated challenges. Neighbor aware solutions can enhance the capabilities of the smart city. In this article, we briefly overview the neighbor aware solutions and challenges in smart cities. We then categorize the neighbor aware solutions and discuss the possibilities using the collaboration among neighbors to extend the lifetime of IoT devices. We also propose a new duty cycle MAC protocol with assistance from the neighbors to extend the lifetime of the nodes. Simulation results further coagulate the impact of neighbor assistance on the performance of IoT devices in smart cities.
- Open Access
- 31 December 2015
- Published by The Korean Institute of Gas in Journal of the Korean Institute of Gas
- Vol. 19 (6) , 40-46
Effect of Operating Condition Change on the Conversion Efficiency of TWC with HCNG Engine
Effect of Operating Condition Change on the Conversion Efficiency of TWC with HCNG Engine HCNG(Hydrogen- Compressed Natural gas Blends);stoichiometric;TWC(Three-way Catalyst);conversion efficiency;equivalence ratio; Stoichiometric combustion engine with Three-way catalyst had an advantage that can reduce the harmful emissions effectively. Fuel equivalence ratio controlled from engine is very important because Fuel equivalence ratio with high conversion efficiency was narrow. This study analyzed the conversion efficiency under whole range of operating area for to evaluate the performance of three-way catalyst. In order to identify the Optimum conversion efficiency, the conversion efficiency due to change the control value of fuel equivalence ratio was investigated. The result show that conversion efficiency of emissions(more than 95%) has discovered by means of fuel equivalence ratio control at each test condition. As engine power increases, optimal fuel equivalence ratio tended to increase linearly under operating conditions of similar exhaust gas temperature.
- 31 December 2015
- Published by The Korean Institute of Gas in Journal of the Korean Institute of Gas
- Vol. 19 (6) , 40-46
Open AccessEffect of Operating Condition Change on the Conversion Efficiency of TWC with HCNG Engine
Effect of Operating Condition Change on the Conversion Efficiency of TWC with HCNG Engine HCNG(Hydrogen- Compressed Natural gas Blends);stoichiometric;TWC(Three-way Catalyst);conversion efficiency;equivalence ratio; Stoichiometric combustion engine with Three-way catalyst had an advantage that can reduce the harmful emissions effectively. Fuel equivalence ratio controlled from engine is very important because Fuel equivalence ratio with high conversion efficiency was narrow. This study analyzed the conversion efficiency under whole range of operating area for to evaluate the performance of three-way catalyst. In order to identify the Optimum conversion efficiency, the conversion efficiency due to change the control value of fuel equivalence ratio was investigated. The result show that conversion efficiency of emissions(more than 95%) has discovered by means of fuel equivalence ratio control at each test condition. As engine power increases, optimal fuel equivalence ratio tended to increase linearly under operating conditions of similar exhaust gas temperature.